A Semg-Controlled Robotic Hand Exoskeleton For Rehabilitation In Post-Stroke Individuals

This study proposed to design a robotic hand exoskeleton for rehabilitation in post-stroke individuals controlled by surface electromyographic (sEMG) signals. A control system consisting of threshold detection, classification module and actuator control was designed. Six representative functional hand gestures was recognized using support vector machine (SVM) based on the sEMG signals captured from eight forearm and hand muscles. Kinematic simulation was performed, by which the maximum joint angles for extension was measured. Results showed that the angle changes in metacarpophalangeal (MCP) joints of index, middle, ring and pinky were more than 7 degrees. The angle changes of proximal interphalangeal (PIP) and distal interphalangeal (DIP) joints of the four fingers exceeded 77 degrees and 43 degrees, respectively. The accuracy of gesture classification was higher than 95.5%, which was superior to the results of previous studies. This device provided an innovative prototype and would play a role in hand rehabilitation in post-stroke individuals.